MCP Server for Iru API (Previously Kandji API)

📢 Rebranding Notice: Kandji has rebranded as Iru. This MCP server continues to work with the Iru platform (formerly Kandji). All API endpoints and functionality remain unchanged. For more information about Iru, visit iru.com.

AI-driven device management through Model Context Protocol. This MCP server enables natural-language interactions with the Iru (formerly Kandji) MDM API via Claude Desktop and other AI platforms.

🚀 Multi-Platform Support: This server is being designed to support multiple AI platforms beyond Claude Desktop, including Google Gemini CLI, Ollama, OpenAI ChatGPT, and Microsoft Copilot Studio. See Multi-Platform Support and the ROADMAP for implementation planning and status.

Features

Core Capabilities

• 23 MCP Tools: Comprehensive device, security, and compliance management
• Device Management: Search, inspect, monitor, and control devices
• Security Monitoring: Vulnerability tracking, threat detection, and behavioral analysis
• Compliance Reporting: Organization-wide compliance and audit logging
• User Management: Directory integration and user lifecycle management
• Configuration Management: Blueprints, tags, and library items

Technical Features

• Smart Caching: TTL-based caching (5min devices, 2min compliance, 30min blueprints)
• Script Generation: Automatic bash scripts for large data exports with pagination handling
• Comprehensive Testing: 456 unit tests with 81% branch coverage (exceeds 80% threshold)
• PII Redaction: Optional privacy mode for sensitive data
• Error Handling: Categorized errors (auth, validation, rate_limit, network, server) with recovery strategies
• Response Envelopes: Standardized response format with tables, metadata, and suggestions
• Security: Explicit confirmation for destructive operations, secure credential storage

Installation

Prerequisites

• Node.js 18 or higher
• Kandji API token
• Kandji subdomain

Setup

1. Clone the repository:

git clone https://github.com/mangopudding/mcp-server-iru-api.git
cd mcp-server-iru-api

2. Install dependencies:

npm install

3. Create .env file from example:

cp .env.example .env

4. Configure your .env file:

KANDJI_API_TOKEN=your_api_token_here
KANDJI_SUBDOMAIN=your_subdomain_here
KANDJI_REGION=us

Getting Your Kandji API Token

1. Log in to your Kandji tenant
2. Navigate to Settings → Access
3. Click "Add API Token"
4. Copy the generated token to your .env file

Verify API Access

After configuring your .env file, verify your API token has the correct permissions:

npm run verify

This will test all API endpoints and report any permission issues. See troubleshooting/README.md for details.

Usage

Development Mode

Run the server in development mode with auto-reload:

npm run dev

Production Mode

Build and run in production:

npm run build
npm start

Testing

Run tests:

npm test

Run tests with coverage:

npm run test:coverage

Configuring Claude Desktop

See config/README.md for detailed setup instructions.

Quick setup:

1. Build the server: npm run build
2. Copy example config: cp config/claude_desktop_config.example.json ~/Desktop/
3. Edit with your Kandji credentials and absolute path
4. Add to ~/Library/Application Support/Claude/claude_desktop_config.json
5. Restart Claude Desktop

Example configuration:

{
  "mcpServers": {
    "kandji": {
      "command": "node",
      "args": ["/absolute/path/to/mcp-server-kandji-api/dist/index.js"],
      "env": {
        "KANDJI_API_TOKEN": "your_token",
        "KANDJI_SUBDOMAIN": "your_subdomain",
        "KANDJI_REGION": "us"
      }
    }
  }
}

For troubleshooting and advanced opti

... View full README on GitHub